Work

DeepColor: Reinforcement Learning optimizes information efficiency and well-formedness in color name partitioning

Year: 2018
Type: article
Abstract: As observed in the World Color Survey (WCS), some univer-sal properties can be identified in color naming schemes overa large number of languages. For example, Regier, Kay, andKhetrapal (2007) and Regier, Kemp, and Kay (2015); Gib-son et al. (2017) recently explained these universal patterns interms of near optimal color partitions and information theoreticmeasures of efficiency of communication. Here, we introducea computational learning framework with multi-agent systemstrained by reinforcement learning to investigate these universalproperties. We compare the results with Regier et al. (2007,2015) and show that our model achieves excellent quantitativeagreement. This work introduces a multi-agent reinforcementlearning framework as a powerful and versatile tool to investi-gate such semantic universals in many domains and contributesignificantly to central questions in cognitive science.
Language: English
Field: Psychology
Open Access status: green
OpenAlex ID: W2907860297
FWCI: 0.603
Cites: 19
Cited by: 4
Related to: 10